CN112069962B - Method for identifying vibration spectrum under strong noise background based on image - Google Patents
Method for identifying vibration spectrum under strong noise background based on image Download PDFInfo
- Publication number
- CN112069962B CN112069962B CN202010885486.7A CN202010885486A CN112069962B CN 112069962 B CN112069962 B CN 112069962B CN 202010885486 A CN202010885486 A CN 202010885486A CN 112069962 B CN112069962 B CN 112069962B
- Authority
- CN
- China
- Prior art keywords
- vibration
- noise
- image
- spectrogram
- spectrum
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000001845 vibrational spectrum Methods 0.000 title claims description 13
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 238000013145 classification model Methods 0.000 claims abstract description 13
- 238000001228 spectrum Methods 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 24
- 238000000605 extraction Methods 0.000 claims description 19
- 238000013527 convolutional neural network Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 7
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
- G01M15/14—Testing gas-turbine engines or jet-propulsion engines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention belongs to the technical field of ground tests of military and civil aviation engines, and can be used for optimizing the vibration test of the whole ground of the aviation engine and reducing the fault removal time. The method is characterized in that the vibration caused by noise signals is large or the vibration caused by mechanical fault signals is large by an image classification technology mode. The method comprises the following steps: 1. and obtaining a spectrogram of the vibration data. 2. The obtained spectrogram is transformed into a 28 x 28 gray scale image. 3. And extracting gray image characteristics through the trained convolution network. 4. And judging whether the spectrogram contains noise or not by taking the extracted gray image features as the input of a classification model. The invention can explore the intelligent classification mode of the vibration test signal of the aeroengine, and reduce the fault removal time.
Description
Technical Field
The invention belongs to the whole machine test technology of an aeroengine, and discloses a method for identifying vibration frequency spectrums under a strong noise background based on images.
Background
When an aeroengine is running, various components can generate complex dynamic information due to complex running conditions. Some of these information is important for monitoring the operation state of the equipment, and dynamic information is often aliased in signals such as noise signals and cannot be effectively identified.
Classification of noise signals and mechanical fault signals is important for fault judgment and handling. And by judging whether the spectrogram contains the confidence coefficient of the noise judgment vibration data, the misjudgment on the engine state is reduced, and the smooth running of the aeroengine test is ensured.
When vibration is bigger, if the diagnosis is that noise signals cause the vibration to be bigger, after parking, test bed technicians only need to do short-time work such as changing damaged cables, newly adding isolated vibration measuring points, moving away radiation interference sources and the like, so that reliable vibration monitoring data can be ensured to be available when testing, and test can be continued. If the diagnosis vibration is large and is a mechanical failure, the relevant professional expert should be informed to carry out analysis and diagnosis. Therefore, the vibration signals are classified by using an image-based classification technology, and whether the vibration caused by noise signals is large or the vibration caused by mechanical fault signals is large is identified.
Disclosure of Invention
The method for identifying the vibration spectrum under the strong noise background based on the image is provided, and the erroneous judgment on the engine state is reduced by judging whether the spectrum diagram contains the confidence of the noise judging vibration data or not, so that the smooth running of the aeroengine is ensured.
No documents are searched for methods related to the frequency spectrum of the noise signal with the large vibration and the frequency spectrum of the mechanical fault signal with the large vibration by using the image classification technique at home and abroad.
The technical scheme is as follows:
a method for identifying vibration spectrum under a strong noise background based on an image comprises the following steps:
step 1 station arrangement
The vibration sensor is arranged at a position where random vibration has less influence;
step 2 Signal processing
And removing trend terms from the acquired original vibration signals by using a segmentation fitting trend term removing method, integrating the signals, and amplifying the amplitude of the low-frequency noise signals so that the characteristics of noise interference signals on the frequency spectrum images are more obvious. Filtering the vibration signal with frequency conversion more than three times by using a low-pass filter, and simplifying the frequency spectrum characteristic on the premise of retaining the pre-three times frequency vibration signal and noise interference; adding a hanning window to the processed front triple frequency vibration signal and noise interference to avoid superposition of leakage phenomenon and noise interference phenomenon; and finally, carrying out frequency spectrum calculation on the front triple frequency vibration signal and noise interference.
Step 3 training of the model
Manufacturing a feature extraction training sample: acquiring a spectrogram according to priori data, and converting the acquired spectrogram into a gray image, wherein the priori data specifically comprises noise data and non-noise data; training a feature extraction model: the gray level image is input into the convolutional neural network, the training label adopts one-hot coding, namely 01 coding, and the output is a full connection layer. Feature extraction: based on the fully connected layer convolutional neural network after training, the training sample is input into the convolutional neural network again, and the output of the previous layer of the fully connected layer is saved as the characteristic of the sample. Classification model: and training a classification model by taking the result of convolutional neural network feature extraction as input.
Step 4, use of the model
The data acquisition device acquires vibration data and calculates a spectrogram in software. And converting the spectrogram into a gray image. And taking the gray level image as input to perform convolutional neural network feature extraction. And judging whether the spectrogram contains noise or not by taking the result of feature extraction as the input of a two-classification model, wherein if the output is 1, the noise amplitude of the vibration test signal is high, and 0 represents the noise amplitude of the vibration test signal is low.
The low pass filter is a Butterworth 4 order filter.
In the signal processing in the step 2, in order to ensure the accuracy of the integration result, the sampling rate needs to be greater than or equal to 5120Hz.
The gray scale image is 28 x 28 pixels.
The spectrogram gray scale image used by the model training is consistent with the spectrogram gray scale image pixel used by the model.
The classification model adopts a support vector machine algorithm.
The spectrum calculation software Matlab, labVIEW.
And 2, taking every 4096 points to perform Fourier transformation when the vibration of the aircraft accessory case is measured.
The beneficial technical effects are as follows: in the prior art, the intelligent classification is realized by using the vibration frequency spectrum and the low-noise vibration frequency under the artificial classification strong noise background. And by judging whether the spectrogram contains the confidence coefficient of the noise judgment vibration data, the misjudgment on the engine state is reduced, and the smooth running of the aeroengine test is ensured.
Drawings
FIG. 1 is a flow chart of vibration noise spectrogram identification.
Detailed Description
The specific flow is shown in figure 1.
The invention provides a method for classifying vibration spectrum and low-noise vibration spectrum under a strong noise background based on an image, which comprises the following steps: according to the method, noise is identified in an image mode, feature extraction is performed through a convolutional neural network, two classification is performed through a support vector machine, and the method for identifying vibration signals with high noise amplitude in an image mode can reduce misidentification of engine states and ensure smooth running of test of an aeroengine.
A method for identifying vibration spectrum under a strong noise background based on an image comprises the following steps:
step 1 station arrangement
The vibration sensor is arranged at a position where random vibration has less influence. Reducing the complexity of the spectrum. And the vibration measuring points are arranged at the positions with better rigidity on the casing, the vibration sensors are arranged near the main mounting joint and the auxiliary mounting joint for evaluating the vibration transmitted by the engine to the airplane. For the vibration of the aircraft accessory case, the vibration sensor cannot be mounted at the cantilever end of the vibration cradle beyond the aircraft accessory case. When the sensor bracket is installed on the engine casing or the aircraft accessory casing, a leveling ruler is used for ensuring the level of the installation surfaces of the sensor bracket and the sensor.
Step 2 Signal processing
And removing trend terms from the acquired original vibration signals by using a segmentation fitting trend term removing method, integrating the signals, and amplifying the amplitude of the low-frequency noise signals so that the characteristics of noise interference signals on the frequency spectrum images are more obvious. Filtering the vibration signal with frequency conversion more than three times by using a low-pass filter, and simplifying the frequency spectrum characteristic on the premise of retaining the pre-three times frequency vibration signal and noise interference; the cut-off frequency of the low-pass filter is set to be 25% higher than the frequency-tripled frequency so as to ensure that the frequency-tripled vibration signal cannot be attenuated obviously. And adding a hanning window to the processed front triple frequency vibration signal and noise interference to avoid superposition of leakage phenomenon and noise interference phenomenon, and if matlab software is used for calculating a frequency spectrum, dividing the windowed signal by an inherent gain for correction. And finally, carrying out frequency spectrum calculation on the front triple frequency vibration signal and noise interference. In order to preserve the noise interference characteristics as much as possible, the spectrum should be selected from an amplitude spectrum or an effective value spectrum, and the power spectrum density cannot be used. The coordinates are unified to linear coordinates, and logarithmic coordinates cannot be used.
Step 3 training of the model
Manufacturing a feature extraction training sample: acquiring a spectrogram according to priori data, and converting the acquired spectrogram into a gray image, wherein the priori data specifically comprises noise data and non-noise data; training a feature extraction model: the gray level image is input into the convolutional neural network, the training label adopts one-hot coding, namely 01 coding, and the output is a full connection layer. Feature extraction: based on the fully connected layer convolutional neural network after training, the training sample is input into the convolutional neural network again, and the output of the previous layer of the fully connected layer is saved as the characteristic of the sample. Classification model: and training a classification model by taking the result of convolutional neural network feature extraction as input.
Step 4 use of the model
The data acquisition device acquires vibration data and calculates a spectrogram in software. And converting the spectrogram into a gray image. And taking the gray level image as input to perform convolutional neural network feature extraction. And judging whether the spectrogram contains noise or not by taking the result of feature extraction as the input of a two-classification model, wherein if the output is 1, the noise amplitude of the vibration test signal is high, and 0 represents the noise amplitude of the vibration test signal is low.
The low pass filter is a Butterworth 4 order filter. The transition between the pass band and the stop band is smooth, and unstable phenomena such as sudden increase and the like can not occur.
In the signal processing in the step 2, in order to ensure the accuracy of the integration result, the sampling rate needs to be greater than or equal to 5120Hz.
The gray scale image is 28 x 28 pixels. The pixel size is selected conventionally in the work, and different pixels can be set according to actual requirements.
The spectrogram gray scale image used by the model training is consistent with the spectrogram gray scale image pixel used by the model. If the two pixels are inconsistent, the final classification result is not ideal.
The classification model adopts a support vector machine algorithm.
The spectrum calculation software Matlab, labVIEW. The two types of software have powerful signal processing functions and short development period.
And 2, taking every 4096 points to perform Fourier transformation when the vibration of the aircraft accessory case is measured. The frequency components of the rotating machinery are more, so that the frequency resolution is improved, and the phenomenon that vibration signals of adjacent frequencies in a frequency spectrum are overlapped and cannot be identified is avoided.
Claims (8)
1. The method for identifying the vibration spectrum under the strong noise background based on the image is characterized by comprising the following steps of:
step 1 station arrangement
The vibration sensor is arranged at a position where random vibration has less influence;
step 2 Signal processing
Removing trend terms from the collected original vibration signals by using a segmentation fitting trend term removing method, then integrating the signals, and amplifying the amplitude of the low-frequency noise signals; filtering the vibration signal with frequency conversion more than three times by using a low-pass filter, and simplifying the frequency spectrum characteristic on the premise of retaining the pre-three times frequency vibration signal and noise interference; adding a hanning window to the processed front triple frequency vibration signal and noise interference to avoid superposition of leakage phenomenon and noise interference phenomenon; finally, carrying out frequency spectrum calculation on the front triple frequency vibration signal and noise interference;
step 3 training of the model
Manufacturing a feature extraction training sample: acquiring a spectrogram according to priori data, and converting the acquired spectrogram into a gray image, wherein the priori data comprises noise data and non-noise data; training a feature extraction model: the gray level image is input into a convolutional neural network, the training label adopts one-hot coding, namely 01 coding, and the output is a full-connection layer; feature extraction: based on the fully-connected layer convolutional neural network after training, re-inputting training samples into the convolutional neural network, and storing the output of the previous layer of the fully-connected layer as the characteristics of the samples; classification model: training a classification model by taking the result of convolutional neural network feature extraction as input;
step 4, use of the model
The data acquisition equipment acquires vibration data and calculates a spectrogram in software; converting the spectrogram into a gray image; taking a gray image as input to carry out convolutional neural network feature extraction; and judging whether the spectrogram contains noise or not by taking the result of feature extraction as the input of a two-classification model, wherein if the output is 1, the noise amplitude of the vibration test signal is high, and 0 represents the noise amplitude of the vibration test signal is low.
2. The method for identifying vibration spectrum in a strong noise background based on image according to claim 1, wherein the low pass filter is a butterworth 4-order filter.
3. The method for recognizing vibration spectrum in strong noise background according to claim 1, wherein the signal processing in step 2 has a sampling rate of 5120Hz or more.
4. The method of claim 1, wherein the gray scale image is 28 x 28 pixels.
5. The method for recognizing vibration spectrum in a strong noise background based on image according to claim 1, wherein the spectrogram gray-scale image used for model training is consistent with the spectrogram gray-scale image pixels used for model.
6. The method for recognizing vibration spectrum in a strong noise background based on image according to claim 1, wherein the classification model adopts a support vector machine algorithm.
7. The method of claim 1, wherein the spectrum calculation software Matlab, labVIEW is configured to identify a spectrum of vibration in a strong noise background based on the image.
8. The method for recognizing vibration spectrum in a strong noise background according to claim 1, wherein in the step 2, when the vibration of the aircraft accessory case is measured, fourier transform is performed every 4096 points.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010885486.7A CN112069962B (en) | 2020-08-28 | 2020-08-28 | Method for identifying vibration spectrum under strong noise background based on image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010885486.7A CN112069962B (en) | 2020-08-28 | 2020-08-28 | Method for identifying vibration spectrum under strong noise background based on image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112069962A CN112069962A (en) | 2020-12-11 |
CN112069962B true CN112069962B (en) | 2023-12-22 |
Family
ID=73660569
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010885486.7A Active CN112069962B (en) | 2020-08-28 | 2020-08-28 | Method for identifying vibration spectrum under strong noise background based on image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112069962B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114136648B (en) * | 2021-10-20 | 2023-06-13 | 中国航发四川燃气涡轮研究院 | Pneumatic excitation identification method for aeroengine fan movable blade based on acoustic array |
CN114112401A (en) * | 2021-11-10 | 2022-03-01 | 中国人民解放军陆军炮兵防空兵学院 | Engine fault diagnosis method of LSTM fault diagnosis model based on spectrogram |
CN117309299B (en) * | 2023-11-28 | 2024-02-06 | 天津信天电子科技有限公司 | Servo driver vibration test method, device, equipment and medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018093444A1 (en) * | 2016-09-07 | 2018-05-24 | Massachusetts Institute Of Technology | High fidelity systems, apparatus, and methods for collecting noise exposure data |
WO2019218725A1 (en) * | 2018-05-16 | 2019-11-21 | 深圳大学 | Intelligent input method and system based on bone-conduction vibration and machine learning |
CN110595780A (en) * | 2019-09-20 | 2019-12-20 | 西安科技大学 | Bearing fault identification method based on vibration gray level image and convolution neural network |
-
2020
- 2020-08-28 CN CN202010885486.7A patent/CN112069962B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018093444A1 (en) * | 2016-09-07 | 2018-05-24 | Massachusetts Institute Of Technology | High fidelity systems, apparatus, and methods for collecting noise exposure data |
WO2019218725A1 (en) * | 2018-05-16 | 2019-11-21 | 深圳大学 | Intelligent input method and system based on bone-conduction vibration and machine learning |
CN110595780A (en) * | 2019-09-20 | 2019-12-20 | 西安科技大学 | Bearing fault identification method based on vibration gray level image and convolution neural network |
Non-Patent Citations (1)
Title |
---|
基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法;李恒;张氢;秦仙蓉;孙远韬;;振动与冲击(19);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN112069962A (en) | 2020-12-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112069962B (en) | Method for identifying vibration spectrum under strong noise background based on image | |
Wang et al. | Fault diagnosis of diesel engine based on adaptive wavelet packets and EEMD-fractal dimension | |
CN110991295B (en) | Self-adaptive fault diagnosis method based on one-dimensional convolutional neural network | |
CN109883706B (en) | Method for extracting local damage weak fault features of rolling bearing | |
CN101936767B (en) | Method for extracting engineering machine running characteristic signals | |
CN109827777A (en) | Rolling bearing fault prediction technique based on Partial Least Squares extreme learning machine | |
CN112378660A (en) | Intelligent fault diagnosis method for aero-engine bearing based on data driving | |
CN110056640B (en) | Speed reducer wireless fault diagnosis method based on acceleration signal and edge calculation | |
CN111077386A (en) | Early fault signal noise reduction method for electrical equipment | |
CN112182490B (en) | Reactor state diagnosis method and system | |
CN114112400A (en) | Mechanical bearing fault diagnosis method based on multi-angle information fusion | |
CN114486263A (en) | Noise reduction and demodulation method for vibration signal of rolling bearing of rotary machine | |
CN112052712A (en) | Power equipment state monitoring and fault identification method and system | |
Zhou et al. | Multi-objective sparsity maximum mode de-composition: a new method for rotating machine fault diagnosis on high-speed train axle box | |
CN112686181B (en) | Hydraulic turbine fault diagnosis method based on interpolation axis track | |
CN211478951U (en) | Fault diagnosis device and fault diagnosis system | |
CN117348093A (en) | Aviation electromagnetic data processing method and system based on ground reference point | |
CN117009870A (en) | Pump cavitation state identification method for frequency domain improved SDP diagram | |
CN117169346A (en) | High-altitude building damage identification method based on wavelet packet energy spectrum analysis | |
CN110287853B (en) | Transient signal denoising method based on wavelet decomposition | |
CN113820133B (en) | Sparse reconstruction method and system for bearing vibration signals | |
CN114781466B (en) | Fault diagnosis method and system based on harmonic fundamental frequency of rotary mechanical vibration signal | |
CN110459197A (en) | Signal Booster and method for faint blind signal denoising and extraction | |
CN109712639A (en) | A kind of audio collecting system and method based on wavelet filter | |
CN113340369B (en) | Signal processing method and device for turbine fuel mass flowmeter |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |